On the Road to Composable – Composable Freedom

There's a movement happening in eCommerce. Something we at Product League call the "road to composable". As you run, build and scale your digital business, you might not even have realised you were on this road. You might think that every problem that comes up with each new eCommerce opportunity is yours alone. Rest assured, we're walking this road together.

Today's topic: Composable Freedom...

This is the third and final part of our blog series exploring the different stages on the road to composable. In the previous article, we looked at the option of going headless, the reasons to consider it, and the value it can deliver. In short: by decoupling your front-end from your back-end, you align your systems more closely with the pace of change. Valuable in itself, and also the first step towards true composability.

Read the earlier articles in our series: "On the Road to Composable – The Monolith" and "On the Road to Composable – Going Headless".

Composable

During your shift to headless, or perhaps even before, you start experimenting with your first microservices. Think of a dedicated on-site search engine or a modern content management system that's headless by nature. You're now genuinely taking the step towards composable: a landscape no longer limited by the question "what still fits within the monolithic platform?", but instead shaped by the question "what fits me best?"

In a composable landscape, your tailored user experience stays decoupled from the back-end (headless). But instead of the back-end being one large, still fairly monolithic e-commerce engine, you start thinking in microservices. Which system, process or service best fits the needs of my business? Investing in a dedicated pricing and promotions engine makes sense if your business relies heavily on pricing and promotion logic to offer customers the best deals at any given moment. An app that uses back-end customer data in exciting ways to surface benefits can set your business apart if you have a custom, tiered loyalty programme.

The revolution begins when you stop seeing your technology as one large back-end system, and instead as a versatile landscape of headless microservices, each specifically designed to meet the goals it needs to fulfil.

In recent years, a movement has emerged, initiated by a number of vendors of this kind of application, called the MACH Alliance. MACH stands for:

• Microservices
• API-first
• Cloud-native
• Headless

These companies realised their products were very good at solving a specific subset of problems. They didn't want to expand and dilute their solutions until their products also became one-size-fits-all platforms. Instead, they chose to focus on becoming best-in-class within their niche.

They decided to collaborate with other companies solving relevant problems and to embrace the principles of MACH architecture to guarantee the best mutual integrations. This way, any (eCommerce) problem can be solved simply by connecting the right puzzle pieces. MACH landscapes are no longer made up of large monolithic platforms, but of multiple specialised services from SaaS vendors that work together seamlessly.

An example of MACH: decoupling a CMS

A classic example of the MACH mindset shift is a dedicated content management system (CMS). An older CMS was responsible for both creating content and rendering it on the site. Often, this type of CMS is still part of a monolithic eCommerce application. The CMS pushes the content out and generates the site according to set rules. This often causes conflicts between the HTML inside the CMS and the rest of the site. It's also difficult to scale, because you have to describe exactly how every component should behave, while accounting for the rest of the monolithic platform.

A headless CMS, by contrast, isn't concerned with rendering content, only with the content itself. For each section of a page, the decoupled front-end asks the CMS: "Do you have content for me?" And the CMS responds with the available data. The front-end then decides for itself how this content is visually displayed.

This "pull" model isn't just much faster than the old "push" model, it also offers far more possibilities. Want to show a banner on the homepage one way and on the search results page another way? No problem, you can use the same content but style it differently. Even better, if you later want to use that content for an in-store display too, you're free to do so. And that's without even mentioning all the personalisation options that let you offer customers relevant, tailored content.

This is the MACH way of working: you communicate with these systems only through APIs. Imagine the relief across your teams: content editors can now create their content without worrying about embedding HTML. They know the rules for the end product and can "fire and forget". Front-end development teams no longer need to worry either, because their carefully designed pages can no longer be broken by a careless or incorrect content update. They retain full control over how it's displayed.

A new all-in-one solution?

Moving from a monolithic and/or headless eCommerce engine to a composable eCommerce landscape certainly brings many benefits. Automatic scalability, so you're never caught off guard by peaks like Black Friday. High performance, because every tool does only what it's meant to do. And a significantly improved time-to-market: Gartner predicted that organisations adopting a composable approach would outpace competitors by 80% in the speed of new feature deployment in 2023.

Instead of one big decision every five years (do we stick with our monolith, upgrade it, or do something else?), you now make many small decisions (is this tool still the best solution for this specific problem?). This means Total Cost of Ownership (TCO) no longer stands in the way of innovation. It's far simpler and cheaper to replace a single microservice than to change your entire eCommerce landscape!

Still, there are certainly challenges on the road to composable. Your vendor landscape becomes more complex. Managing all these contracts and vendors can be quite challenging. Also important: is your organisation able to orchestrate this kind of digital transformation? It requires strict adherence to architectural principles, a much higher demand on teams to break down silos, and an entirely new organisational mindset. And let's not forget that the technical skills needed to build and manage this new landscape, with its new data needs, API orchestration and custom front-end flows, shouldn't be underestimated.

There are two approaches to tackling this complexity:

1. You try to buy your way out of this problem by purchasing as many SaaS solutions as possible, with implementation partners for all of them. The problem here is that it still requires a lot of coordination between all these different partners. What's more, SaaS solutions, especially on the front-end side, offer a lot of functionality out of the box, but are limited in customisation options, and the code is often hard to reuse.
2. You try to build your way out of this problem by deploying large teams of engineers who build custom solutions with high code. While this certainly delivers the most flexible and best-performing applications, it requires an organisation to become an IT company, which for many retailers is a distant dream. What's more, this route can become the most expensive if you try to build every solution yourself that other companies have already figured out for you.

Build better experiences with composable commerce + low code

In practice, most organisations choose a combination of both strategies: they buy solutions for "solved" problems (such as commercetools for the e-commerce engine or Contentful as a CMS) and invest in custom solutions where the most value is to be found. This custom work is either developed in-house, or in collaboration with external development partners.

We see most of the growth in the eCommerce solutions market happening between these Buy and Build tactics. On the Buy side, SaaS solutions are becoming increasingly flexible and geared towards composable architectures. This applies to many modern, headless SaaS solutions, but also to older monoliths moving to the cloud to stay connected with younger competitors (though only time will tell whether these are viable solutions or just a form of "cloud washing"). A major downside of these SaaS solutions, however, remains their lack of flexibility. What if you need a custom front-end for a specific business use case? Or an orchestration service for data between a few systems that aren't used often, but are crucial to your digital experiences? These use cases come up often in an eCommerce landscape, but without investing in custom development, they slowly slip to the bottom of your backlog.

On the Build side, we see a rise of low-code platforms. Low code focuses on making custom development more accessible by lowering costs. Either you need fewer developers for the same result, or the same number of developers delivers more. Some eCommerce front-end SaaS solutions, such as commercetools frontend, are moving in this direction, enabling citizen developers to build solutions themselves. Even Salesforce offers a very limited low-code drag-and-drop eCommerce starter kit. There are also pure low-code platforms entering this market.
At Product League, we choose the high-performance low-code platform OutSystems to develop our eCommerce solutions. We chose OutSystems because it embraces reusability. You can build something once and publish it both as a responsive website and as a hybrid app. You can create integration modules that make it faster to integrate systems, which becomes increasingly important as your landscape shifts towards composable. You can also quickly build multiple front-ends for specific back-end users. Because eCommerce is central to your business process, many users need access to eCommerce systems with specific functionality that can't easily be served by one generic "backend portal". With OutSystems, we can build both customer-facing front-ends and specific front-ends for the employees managing eCommerce, on the same back-end.

The downside of the Build approach remains the lack of reusable solutions. After all, you don't want to keep reinventing the wheel. Unfortunately, you often spend the first few months rebuilding what SaaS solutions have already done before, before you can even start on the real custom work. This is where the exciting concept of Accelerators comes in. An accelerator is designed to speed up your custom development with out-of-the-box components. This lets you build on prior work, so a project starts at 60% instead of 0%. At Product League, we've chosen this approach because it fits seamlessly with the low-code mindset: putting reusability at the centre of designing and building every component. By keeping reusability and accelerator-driven development in mind, you start thinking early about where to buy and where to build. This lets you integrate "solved" solutions into your architecture and focus your development efforts on the use cases that truly set you apart.

A journey, not a destination

Where does your business stand on the road to composable? We've been through many peaks, valleys and bumps in this 3-part blog series. Here's a list of the questions we've asked ourselves throughout this series:

• Does my eCommerce technology meet customer demand, without scaling issues?
• Is my technology stack holding me back?
• What business problems make my company unique?
• What unique value do I offer customers?
• Am I trying to solve new problems, or am I running into existing ones?

As you answer these questions yourself, we hope you leave this series with a sense of direction and purpose, to apply the principles of composable where they make sense, and to set them aside where they don't (yet) offer a good solution.

Building a new eCommerce legacy

Our 'Road to Composable' series may have come to an end, but your journey towards eCommerce excellence might just be getting started. It's time to turn insights into action and theories into tangible results, and working with Product League will help you explore the exciting world of composable eCommerce. Don't just adapt to change, be the one leading it. Get in touch now, and together we'll create a new legacy in eCommerce by building systems as unique and dynamic as your business vision!